A comprehensive evaluation method for the competitiveness of charging stations

Abstract

This invention discloses a comprehensive evaluation method for the competitiveness of charging stations, comprising the following steps: determining the final set of indicators for evaluating station competitiveness; objectively assigning weights to each indicator using the entropy weight method based on an indicator weight model; constructing a comprehensive evaluation model one for station competitiveness based on grey relational analysis; constructing a comprehensive evaluation model two for station competitiveness based on TOPSIS analysis; ranking and comparing the station competitiveness indices obtained from Model 1 and Model 2, cross-validating the results to obtain their validity, and obtaining the final ranking of station competitiveness through an averaging algorithm; and visualizing the target station competitiveness data. This invention's comprehensive evaluation method for charging station competitiveness has the advantages and characteristics of accessible indicator data, cross-validation of the dual evaluation models, and full consideration of the competitive environment surrounding the charging station. Combined with the subsequent matching visualization scheme for the evaluation results, this evaluation method has strong practical applicability.

Description

Technical Field

[0001] This invention relates to a comprehensive evaluation method for the competitiveness of charging stations. Background Technology

[0002] In the face of increasingly fierce competition in the charging market, self-operated charging stations need precise service positioning and efficient pricing strategies to enhance their competitiveness. Charging station operators require a reasonable competitiveness analysis model to analyze the basic data of target stations and surrounding stations to assess their competitiveness and provide guidance for subsequent self-operated station management and operational optimization. The evaluation model mainly consists of two parts: evaluation indicators and evaluation methods.

[0003] Many domestic studies have evaluated the development of service-related industries by constructing multi-indicator competitiveness assessment models. For example, Zhang Zhong and Jin Qing (2015) selected 16 indicators from four perspectives—strategy, process, performance, and customer value—to construct an evaluation index system for service-oriented manufacturing, and used a forward-feedback three-layer BP neural network as the evaluation model. Bian Guoli (2017) selected 6 secondary indicators and 12 tertiary indicators from three perspectives—basic factors, industry factors, and market factors—and evaluated the service outsourcing industry based on the grey AHP method.

[0004] For the specific object of electric vehicle charging stations, there are also some related studies, including: Xie Menghua (2021) set 12 indicators from 4 dimensions: integrated planning and operation indicators, distribution network operation impact indicators, transportation network operation impact indicators, and user experience indicators. These indicators include the number of charging piles, maintenance costs, traffic operation index, average charging time, average queuing time, and parking fee level. The Analytic Hierarchy Process (AHP) was then used for calculation and evaluation. Dong Hua (2023) selected 19 indicators from 4 aspects: planning service, operator service, user service, and social service. A comprehensive evaluation index system for electric vehicle charging station services was constructed. A combined weighting method was then used, employing subjective weighting methods (AHP method, ordinal relation method) and objective weighting methods (entropy weight method, coefficient of variation method) to calculate the weights. Finally, the comprehensive weights were determined based on the moment estimation theory. Liu Si (2016) proposed a method for setting electric vehicle charging prices by comprehensively applying cost-benefit analysis, starting from the factors affecting the operating efficiency and consumer benefits of charging stations. Zhang Wei (2017) selected 14 indicators from three perspectives: operational status, customer service, and impact on traffic and power distribution network, and constructed a combined evaluation model based on three methods: the "average method", the "Borda method", and the "Copeland method".

[0005] However, considering their application in real-world scenarios of evaluating the competitiveness of charging stations and adjusting service pricing, the aforementioned models suffer from three core problems: difficulty in obtaining indicator data, lack of regional competition consideration in indicator selection, and a lack of suitable data visualization methods. These issues hinder their practical application. First, existing models have excessively high requirements for the calculated data, and the data acquisition methods are not universally applicable, making it difficult to obtain data from public platforms. This results in high data requirements and hinders widespread application. Second, different charging stations have varying basic operating conditions and regional competitive environments. The existing charging station operation service evaluation models and indicator systems lack analysis of regional competition, specifically, a comparative analysis of the pricing and operating conditions of other competing stations within the region. Therefore, the evaluation models cannot provide a relative competitiveness assessment for the target station and its competitive environment, offering no guidance for subsequent pricing adjustments and service improvements. Third, the data presented by most models is highly complex, presenting significant analytical challenges, and lacks effective visualization methods to showcase the results for application in practical management. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a comprehensive evaluation method for the competitiveness of charging stations. Based on the entropy weight method, grey relational analysis and TOPSIS method, the method comprehensively evaluates the competitiveness of charging stations. It has the advantages and characteristics of accessible indicator data, cross-validation of dual evaluation models, and full consideration of the competitive environment around the charging station. In addition, the subsequent matching evaluation result visualization scheme makes the evaluation method highly applicable in practice.

[0007] The technical solution to achieve the above objectives is: a comprehensive evaluation method for the competitiveness of charging stations, comprising the following steps:

[0008] S1, determine the final set of indicators for evaluating the competitiveness of the site;

[0009] S2, based on the indicator weight model, uses the entropy weight method to objectively assign weights to each indicator;

[0010] S3, A comprehensive evaluation model for the competitiveness of a site is constructed based on grey relational analysis.

[0011] S4. Based on the TOPSIS analysis method (Technique for Order Preference by Similarity to an Ideal Solution, also known as the superior-inferior solution distance method), a second comprehensive evaluation model for the competitiveness of the site is constructed to calculate the competitiveness index and cross-validate it with the first comprehensive evaluation model for the competitiveness of the site.

[0012] S5. The station competitiveness indices obtained from the comprehensive evaluation model 1 and the comprehensive evaluation model 2 of station competitiveness are ranked and compared. The validity of the results is verified by cross-validation, and the final ranking of station competitiveness is obtained by averaging algorithm.

[0013] S6. Visualize the target site competitiveness data; based on the grey relational coefficient calculation, obtain the correlation coefficient between each indicator data of the object and the optimal value, convert it into a percentage score, and draw a site competitiveness radar chart as a reference for site competitiveness improvement analysis.

[0014] In the above-mentioned comprehensive evaluation method for the competitiveness of charging stations, step S1 involves screening and integrating the existing set of evaluation indicators, and adding regional competition-related indicators to obtain a preliminary indicator set containing 18 indicators. The preliminary indicator set is shown in Table 1.

[0015] Table 1. Preliminary set of indicators integrating existing charging station evaluation indicators:

[0016]

[0017] Table 2, Final set of indicators after filtering based on data accessibility:

[0018]

[0019] Based on the goal of improving the operational status of the facilities and regional competitiveness, redundant indicators that have overlapping effects with each other in the initial selection of indicators were eliminated. The final set of indicators for evaluating the competitiveness of the facilities was determined based on the data fields publicly available on the platform to ensure data accessibility. The final set of indicators contains 9 indicators, as shown in Table 2.

[0020] The above-mentioned comprehensive evaluation method for the competitiveness of charging stations, in step S2, the calculation steps of objectively assigning weights to each indicator using the entropy weight method based on the indicator weight model include:

[0021] S21, Construct the evaluation index matrix;

[0022] S22, Standardization of indicator data;

[0023] S23, indicator data normalization;

[0024] S24, calculate the information entropy of each indicator;

[0025] S25, determine the weight of each indicator.

[0026] The above-mentioned comprehensive evaluation method for the competitiveness of charging stations includes the following calculation steps:

[0027] S21, Construct the evaluation index matrix: Assuming there are m evaluation objects and n evaluation indicators, construct the evaluation index matrix X based on the actual data of each indicator:

[0028]

[0029] S22, Standardization of Indicator Data: There are three ways to evaluate indicator values ​​and judge their merits based on their characteristics:

[0030] When the indicator value is as high as possible, it is considered a positive indicator, and its standardized formula is:

[0031]

[0032] When a smaller indicator value is better, then the indicator is considered an inverse indicator, and its standardized formula is:

[0033]

[0034] The closer the indicator value is to a certain value, the better the evaluation result; this value is denoted as... This indicator is a suitability indicator, and its standardized formula is:

[0035]

[0036] In formulas (2-2), (2-3), and (2-4), x ij This represents the value of the i-th indicator for the j-th evaluation object:

[0037] The matrix X is processed according to the standardization formulas to obtain the normalized matrix X. * :

[0038]

[0039] S23, indicator data normalization, the normalization formula is:

[0040]

[0041] According to the normalization formula (2-6), for matrix X * After normalization, the standardized matrix F is obtained:

[0042]

[0043] S24, calculate the information entropy of each indicator:

[0044]

[0045] In equation (2-8), K takes the value of H i It is the information entropy of the j-th indicator, when f ij When = 0, f ij ln f ij=0; Determine the entropy weights of each indicator:

[0046]

[0047] S25, Determine the weights of each indicator: After the above calculation steps, the entropy weight vector W = (w1, w2, ..., w) of the n evaluation indicators can be obtained. n ) T ,in

[0048] The above-mentioned comprehensive evaluation method for charging station competitiveness, in step S3, includes the calculation steps for constructing the comprehensive evaluation model for charging station competitiveness based on grey relational analysis:

[0049] S31, Establish the evaluation index matrix;

[0050] S32, Construct a comprehensive evaluation model based on grey relational analysis;

[0051] S33, determine the optimal set of indicators;

[0052] S34, Standardization of the indicator matrix;

[0053] S35, Calculate the grey relational coefficient;

[0054] S36, calculate and sort the grey relational degree.

[0055] The above-mentioned comprehensive evaluation method for the competitiveness of charging stations includes the following calculation steps:

[0056] S31, Establish the evaluation index matrix: First, assuming there are m evaluation objects (electric vehicle charging stations) and n evaluation indicators, first establish the evaluation index matrix X, x ij This represents the value of the i-th indicator for the j-th evaluation object:

[0057]

[0058] The comprehensive evaluation model based on grey relational analysis is as follows:

[0059] P = E × W (3-2)

[0060] In equation (3-2), P = (p1, p2, ..., p... m ) T Let p be the vector of comprehensive evaluation results for m evaluation objects. j This represents the grey relational degree of the j-th evaluation object;

[0061] W = (w1, w2, ..., w n ) T Let E be the weight allocation vector for n evaluation indicators, and E be the grey relational coefficient matrix.

[0062]

[0063] In equation (3-3), ε ij ε is the correlation coefficient between the i-th indicator and the i-th optimal indicator of the j-th evaluated object, representing the degree of correlation between the data columns of each evaluation indicator and the optimal data column; ij The larger the value, the greater the correlation between the two data sequences on the i-th indicator;

[0064] Finally, according to p j The numerical values ​​are used to rank the evaluated objects.

[0065] S32, Determine the optimal index set: Assumption The optimal sequence is composed of the optimal values ​​selected from each indicator data column. The selection principle varies depending on the nature of the evaluation indicator. The selection principle is as follows: if an evaluation indicator is a positive indicator, that is, the larger the indicator value, the better, then the maximum value among all evaluated objects is selected; conversely, if an evaluation indicator is an inverse indicator, that is, the smaller the indicator value, the better, then the minimum value among all evaluated objects is selected; if an evaluation indicator is a moderate indicator, that is, when it is good to be close to a certain suitable value, then this moderate value is selected as the element of the reference sequence.

[0066] After determining the optimal index set, construct matrix D:

[0067]

[0068] In equation (3-4), x ij This represents the i-th indicator of the j-th evaluation object. This represents the optimal value of the i-th indicator;

[0069] S33, Standardization of Indicator Matrix D: During the evaluation process, the different dimensions of each evaluation indicator can affect the evaluation results. It is necessary to perform dimensionless quantification on the evaluation indicators, transforming the values ​​of each indicator into a relatively uniform measure. This process is called evaluation indicator data standardization; assuming X = (x ij ) m×n After data standardization, we get R = (r ij ) m×n r ij ∈[0,1]; The standardization process using formulas (2-2), (2-3), and (2-4) involves three cases:

[0070] When the indicator value is as high as possible, it is considered a positive indicator, and its standardized formula is:

[0071]

[0072] When a smaller indicator value is better, then the indicator is considered an inverse indicator, and its standardized formula is:

[0073]

[0074] The closer the indicator value is to a certain value, the better the evaluation result; this value is denoted as... This indicator is a suitability indicator, and its standardized formula is:

[0075]

[0076] The normalized matrix R is obtained by processing matrix X according to the standardization formula:

[0077]

[0078] S34, Calculate the grey relational coefficient: Relational coefficient ε ij The calculation formula is:

[0079]

[0080] In equation (3-9), ρ is the resolution coefficient, which is between 0 and 1. The resolution coefficient is introduced to reduce the influence of the extreme values ​​of the evaluation index on the calculation, thereby improving the discrimination of the correlation coefficient. The smaller the resolution coefficient, the greater the resolution. In general, the resolution coefficient is usually taken as 0.5.

[0081] S35, Calculating Grey Relational Degree: The correlation coefficient compares the degree of correlation between the sequences of each scheme and the optimal sequence on a certain indicator, but it cannot fully reflect the merits of the schemes. In order to grasp the degree of correlation between the sequences of schemes as a whole, it is necessary to weight and accumulate the correlation coefficients into a single number, namely the correlation degree p. Then, the correlation degree of the comparison sequence with respect to the optimal sequence is:

[0082]

[0083] In equation (3-10), w i The weight of the i-th indicator in the evaluation indicator system;

[0084] S36, Ranking: Grey Relational Degree p j The larger the value, the closer the sequence of objects is to the optimal sequence, indicating that the j-th evaluated object is better. Therefore, based on the grey relational degree p... j The size will be used to sort the objects being evaluated.

[0085] In the above-mentioned comprehensive evaluation method for the competitiveness of charging stations, step S4 involves TOPSIS analysis. The principle of TOPSIS analysis is to determine the standardized decision matrix by standardizing the evaluation index data and the weights of each index, and then find the optimal and worst solutions, i.e., positive and negative ideal solutions, in the standardized decision matrix. The distance between each evaluation solution and the positive and negative ideal solutions is calculated respectively, and finally the relative closeness of each solution to the optimal solution is used to evaluate the merits of the solution.

[0086] The above-mentioned comprehensive evaluation method for charging station competitiveness, based on the TOPSIS analysis method, includes the following calculation steps for constructing the second comprehensive evaluation model for charging station competitiveness:

[0087] S41, Establish the evaluation index matrix;

[0088] S42, Standardization of evaluation indicators;

[0089] S43, determine the entropy weight;

[0090] S44, Construct the decision matrix;

[0091] S45, the positive and negative ideal solutions are determined by the optimal value of the matrix V index;

[0092] S46, calculate the Euclidean distance between each evaluation object and the positive and negative ideal solutions;

[0093] S47, calculate the relative proximity to the optimal object.

[0094] The above-mentioned comprehensive evaluation method for the competitiveness of charging stations includes the following calculation steps:

[0095] S41, Establish the evaluation index matrix: First, assuming there are m evaluation objects (electric vehicle charging stations) and n evaluation indicators, establish the evaluation index matrix X, x ij This represents the value of the i-th indicator for the j-th evaluation object:

[0096]

[0097] S42, Standardization of evaluation indicators: Refer to formulas (2-2), (2-3), and (2-4) to standardize the evaluation indicator matrix.

[0098] When the indicator value is as high as possible, it is considered a positive indicator, and its standardized formula is:

[0099]

[0100] When a smaller indicator value is better, then the indicator is considered an inverse indicator, and its standardized formula is:

[0101]

[0102] The closer the indicator value is to a certain value, the better the evaluation result; this value is denoted as... This indicator is a suitability indicator, and its standardized formula is:

[0103]

[0104] The normalized matrix R is obtained by processing matrix X according to the standardization formula:

[0105]

[0106] S43, Determine the entropy weight: Referring to the steps of the entropy weight method, calculate the entropy weight matrix W = (w1, w2, ..., w3) using formula (2-9). n ) T ,in

[0107] S44, Constructing the decision matrix V: Constructing the decision matrix v based on evaluation indicators and entropy weights. ij =r ij ×w i :

[0108]

[0109] S45, determine the positive and negative ideal solutions Z based on the optimal values ​​of each evaluation index in the decision matrix. + and Z - :

[0110]

[0111] S46, Calculate the Euclidean distance D between each evaluation object and the positive and negative ideal solutions. j + and D j -

[0112]

[0113]

[0114] S47, Calculate the relative proximity C between each evaluation object and the optimal object. j :

[0115]

[0116] In equation (4-11), C j The larger the value, the closer the j-th evaluation object is to the optimal object level.

[0117] In the above-mentioned comprehensive evaluation method for the competitiveness of charging stations, in step S5, the TOPSIS analysis method calculates the weighted Euclidean distance and relative proximity, and sorts them according to the relative proximity; while the principle of the grey relational analysis method is to calculate the correlation degree and sort the evaluation objects according to the correlation degree. The two models yield different results, and the final ranking of the station competitiveness is obtained through the averaging algorithm.

[0118] The comprehensive evaluation method for the competitiveness of charging stations of the present invention has the following advantages compared with the prior art:

[0119] (1) The evaluation index system in this invention is an integration, screening, and redundancy removal of existing index systems, selecting 9 indicators in 3 directions for evaluation. All index data can be obtained from publicly available data on online charging platforms, ensuring data accessibility;

[0120] (2) This invention uses grey relational analysis and TOPSIS superior-inferiority distance method to evaluate index data, which can cross-validate the ranking results, and the final ranking result is the average of the two, reducing the one-sidedness of a single evaluation method.

[0121] (3) The selection of indicators in this invention fully considers the competitive environment surrounding the target station, including the number of competing stations in the vicinity and the pricing of service fees. This makes the evaluation model more targeted to the target station, and the evaluation results are more valuable for actual station adjustments;

[0122] (4) By converting the score through grey relational coefficient and visualizing radar chart data, the evaluation results can be presented more clearly and effectively show the advantages and disadvantages of the target site on multiple competitiveness evaluation indicators. Attached Figure Description

[0123] Figure 1 A flowchart of the charging station competitiveness comprehensive evaluation method of the present invention;

[0124] Figure 2 This is an example of calculating the competitiveness of a site based on a dual model using the grey relational analysis method and the TOPSIS method.

[0125] Figure 3 A radar diagram showing the competitiveness indicators of target site A and comparison sites B / F. Detailed Implementation

[0126] To enable those skilled in the art to better understand the technical solution of the present invention, its specific embodiments are described in detail below with reference to the accompanying drawings:

[0127] Please see Figure 1 , Figure 2 and Figure 3In this embodiment of the invention, data is first requested from the databases of mainstream electric vehicle charging platforms in Shanghai (such as Lianlian Charging Pro and e-Charge). After deduplication and error removal, basic information on 10,085 charging stations in Shanghai is obtained, encompassing approximately 173,000 DC and AC charging piles. Important charging station information obtainable from the charging platforms includes fields such as: electricity cost, service fee, total charging cost, number of fast charging piles, number of slow charging piles, parking fee, location tag, parking discount tag, and station level tag.

[0128] A charging station operated by State Grid Electric Vehicles in a certain southern area of ​​the city (Shanghai Minhang District B&Q Electric Vehicle Charging Station) was selected as the target site A* and location benchmark for competitiveness evaluation. Ten surrounding charging stations were selected as the objects of competitiveness evaluation. The peak electricity price on September 18 (8:00-15:00) was selected as the cost and time benchmark. After data processing and qualitative indicator quantification, the basic data are shown in Table 3 below.

[0129] Table 3. Data on a State Grid electric vehicle station and its surrounding competing stations:

[0130]

[0131]

[0132] After obtaining the example data, substitute it into the entropy weight method to calculate the weight of each indicator, and the indicator weight matrix can be obtained, as shown in Table 4.

[0133] Table 4. Calculation results of the entropy weight method example:

[0134] index Information entropy e Variation index d Entropy weight w Fast charging 0.7375 0.2625 0.1616 Slow fill 0.9399 0.0601 0.0370 7-day charging cycles 0.7935 0.2065 0.1271 Total cost (peak) 0.9363 0.0637 0.0392 Parking Fee 0.2598 0.7402 0.4555 Site rating 0.8824 0.1176 0.0724 Service discounts 0.9543 0.0457 0.0282 Number of stations within 2km 0.9352 0.0648 0.0399 Lowest price difference in the surrounding area 0.9363 0.0637 0.0392

[0135] Then, substituting the values ​​into the grey relational analysis model (Model 1) and the TOPSIS model (Model 2) respectively, near-optimality calculations were performed. This yielded the near-optimality calculation values ​​for all stations under both models, as well as their competitiveness ranking among the 10 competing stations. Finally, using the two sets of ranked data and an averaging algorithm, the final ranking of station competitiveness was obtained, as shown below. Figure 2 See Table 5. The algorithm results show that the two algorithms produce similar results, and the sorting accuracy is high.

[0136] Table 5. Ranking results of site competitiveness calculated by grey relational analysis and TOPSIS method:

[0137] station Grey relational degree p Sort 1 station relative proximity c Sort 2 Final sorting A 0.4363 7 A 0.0932 7 7 B 0.4012 10 B 0.0745 10 10 C 0.4670 5 C 0.1079 6 5 D 0.4198 9 D 0.1650 2 5 E 0.4210 8 E 0.0812 9 9 F 0.7028 1 F 0.3407 1 1 G 0.4963 4 G 0.1143 5 4 H 0.4445 6 H 0.0825 8 7 I 0.5498 2 I 0.1647 3 2 J 0.5297 3 J 0.1491 4 3

[0138] Finally, a grey relational analysis was conducted on the target site using nine indicators, and the results were converted into a percentage-based score to visually display the site's competitiveness across various dimensions. A radar chart was also generated. (See attached image.) Figure 3 See Table 6.

[0139] Table 6. Grey relational coefficients and score transformations for each indicator of target station A and comparison stations B / F:

[0140] Target station A score B score F score Fast charging 0.4706 47 0.4211 42 0.3333 33 Slow fill 0.6000 60 0.3333 33 1.0000 100 7-day charging cycles 0.3353 34 0.4225 42 0.3333 33 Total cost (peak) 0.5024 50 0.5261 53 0.3333 33 Parking Fee 0.3333 33 0.3333 33 1.0000 100 Site rating 0.5000 50 0.3333 33 0.5000 50 Service discounts 1.0000 100 1.0000 100 1.0000 100 Number of stations within 2km 1.0000 100 0.5472 55 0.5918 59 Lowest price difference in the surrounding area 0.5024 50 0.5261 53 0.3333 33

[0141] Based on the overall ranking of target site A (B&Q Electric Vehicle Charging Station in Minhang District, Shanghai) among surrounding sites and the comparison of scores across nine key indicators, it can be concluded that site A ranks relatively low (7th) in the competition among surrounding sites, indicating weak overall competitiveness. Stronger competitive indicators include the number of competing sites in the surrounding area, additional services, and preferential offers. To further enhance the site's competitiveness, measures should be considered from the perspectives of parking fees, the number of charging sessions within 7 days, the number of fast charging piles, total charging costs, the price difference with the lowest-priced site in the surrounding area, and site rating.

[0142] In summary, the comprehensive evaluation method for charging station competitiveness of the present invention is based on the entropy weight method, grey relational analysis and TOPSIS analysis to comprehensively evaluate the competitiveness of charging stations. It has the advantages and characteristics of accessible indicator data, cross-validation of dual evaluation models, and full consideration of the competitive environment around the charging station. In addition, the subsequent matching evaluation result visualization scheme makes the evaluation method highly applicable in practice.

[0143] Those skilled in the art should recognize that the above embodiments are merely illustrative of the present invention and are not intended to limit the present invention. Any variations or modifications to the above embodiments that are within the spirit and essence of the present invention will fall within the scope of the claims of the present invention.

Claims

1. A comprehensive evaluation method for the competitiveness of charging stations, characterized in that, Includes the following steps: S1, determine the final set of indicators for evaluating the competitiveness of the site; S2, based on the indicator weight model, uses the entropy weight method to objectively assign weights to each indicator; S3, A comprehensive evaluation model for the competitiveness of a site is constructed based on grey relational analysis. S4, Constructing a comprehensive evaluation model for station competitiveness based on TOPSIS analysis; S5 compares and ranks the station competitiveness indices obtained from the comprehensive evaluation models 1 and 2, and cross-validates the validity of the results. Specifically, the TOPSIS analysis method calculates the weighted Euclidean distance and relative proximity, and ranks the stations based on the relative proximity; while the grey relational analysis method calculates the correlation degree and ranks the evaluation objects based on the correlation degree. The two models yield different results, and the final ranking of station competitiveness is obtained through an averaging algorithm. S6. Visualize the target site competitiveness data; based on the grey relational coefficient calculation, obtain the correlation coefficient between each indicator data of the object and the optimal value, convert it into a percentage score, and draw a site competitiveness radar chart as a reference for site competitiveness improvement analysis.

2. The comprehensive evaluation method for the competitiveness of charging stations according to claim 1, characterized in that, In step S1, after screening and integrating the existing set of evaluation indicators, regional competition-related indicators are added to obtain a preliminary indicator set containing 18 indicators, as shown in Table 1: Table 1. Preliminary set of indicators integrating existing charging station evaluation indicators: Based on the goal of improving the operational status of the facilities and regional competitiveness, redundant indicators that had overlapping influences were eliminated from the initial selection of indicators. A final set of indicators for evaluating facility competitiveness was determined based on publicly available data fields from the platform to ensure data accessibility. The final indicator set contains nine indicators, as shown in Table 2. Table 2, Final set of indicators after filtering based on data accessibility: 。 3. The comprehensive evaluation method for the competitiveness of charging stations according to claim 1, characterized in that, In step S2, the calculation steps for objectively assigning weights to each indicator using the entropy weight method based on the indicator weight model include: S21, Construct the evaluation index matrix; S22, Standardization of indicator data; S23, indicator data normalization; S24, calculate the information entropy of each indicator; S25, determine the weight of each indicator.

4. The comprehensive evaluation method for the competitiveness of charging stations according to claim 3, characterized in that, Each calculation step is as follows: S21, Construct the evaluation index matrix: Assuming there are m evaluation objects and n evaluation indicators, construct the evaluation index matrix X based on the actual data of each indicator: S22, Standardization of Indicator Data: There are three ways to evaluate indicator values ​​and judge their merits based on their characteristics: When the indicator value is as high as possible, it is considered a positive indicator, and its standardized formula is: When a smaller indicator value is better, then the indicator is considered an inverse indicator, and its standardized formula is: The closer the indicator value is to a certain value, the better the evaluation result; this value is denoted as... This indicator is a suitability indicator, and its standardized formula is: In formulas (2-2), (2-3), and (2-4), x ij This represents the value of the i-th indicator for the j-th evaluation object: The matrix X is processed according to the standardization formulas to obtain the normalized matrix X. * : S23, indicator data normalization, the normalization formula is: According to the normalization formula (2-6), for matrix X * After normalization, the standardized matrix F is obtained: S24, calculate the information entropy of each indicator: In equation (2-8), K takes the value of H i It is the information entropy of the j-th indicator, when f ij When = 0, f ij lnf ij =0; Determine the entropy weights of each indicator: S25, Determine the weights of each indicator: After the above calculation steps, the entropy weight vector W = (w1, w2, ..., w) of the n evaluation indicators can be obtained. n ) T ,in 5. The method for comprehensive evaluation of the competitiveness of charging stations according to claim 1, characterized in that, In step S3, the calculation steps for constructing the comprehensive evaluation model of station competitiveness based on grey relational analysis include: S31, Establish the evaluation index matrix; S32, Construct a comprehensive evaluation model based on grey relational analysis; S33, determine the optimal set of indicators; S34, Standardization of the indicator matrix; S35, Calculate the grey relational coefficient; S36, calculate and sort the grey relational degree.

6. The method for comprehensive evaluation of the competitiveness of charging stations according to claim 5, characterized in that, The specific calculation steps are as follows: S31, Establish the evaluation index matrix: First, assuming there are m evaluation objects and n evaluation indicators, first establish the evaluation index matrix X, x ij This represents the value of the i-th indicator for the j-th evaluation object: The comprehensive evaluation model based on grey relational analysis is as follows: In equation (3-2), P = (p1, p2, ..., p... m ) T Let p be the vector of comprehensive evaluation results for m evaluation objects. j This represents the grey relational degree of the j-th evaluation object; W = (w1, w2, ..., w n ) T Let E be the weight allocation vector for n evaluation indicators, and E be the grey relational coefficient matrix. In equation (3-3), ε ij ε is the correlation coefficient between the i-th indicator and the i-th optimal indicator of the j-th evaluated object, representing the degree of correlation between the data columns of each evaluation indicator and the optimal data column; ij The larger the value, the greater the correlation between the two data sequences on the i-th indicator; Finally, according to p j The numerical values ​​are used to rank the evaluated objects. S32, Determine the optimal index set: Assumption The optimal sequence is composed of the optimal values ​​selected from each indicator data column. The selection principle varies depending on the nature of the evaluation indicator. The selection principle is as follows: if an evaluation indicator is a positive indicator, that is, the larger the indicator value, the better, then the maximum value among all evaluated objects is selected; conversely, if an evaluation indicator is an inverse indicator, that is, the smaller the indicator value, the better, then the minimum value among all evaluated objects is selected; if an evaluation indicator is a moderate indicator, that is, when it is good to be close to a certain suitable value, then this moderate value is selected as the element of the reference sequence. After determining the optimal index set, construct matrix D: In equation (3-4), x ij This represents the i-th indicator of the j-th evaluation object. This represents the optimal value of the i-th indicator; S33, Standardization of Indicator Matrix D: During the evaluation process, the different dimensions of each evaluation indicator can affect the evaluation results. It is necessary to perform dimensionless quantification on the evaluation indicators, transforming the values ​​of each indicator into a relatively uniform measure. This process is called evaluation indicator data standardization; assuming X = (x ij ) m×n After data standardization, we get R = (r ij ) m×n r ij ∈[0,1]; The standardization process using formulas (2-2), (2-3), and (2-4) involves three cases: When the indicator value is as high as possible, it is considered a positive indicator, and its standardized formula is: When a smaller indicator value is better, then the indicator is considered an inverse indicator, and its standardized formula is: The closer the indicator value is to a certain value, the better the evaluation result; this value is denoted as... This indicator is a suitability indicator, and its standardized formula is: The normalized matrix R is obtained by processing matrix X according to the standardization formula: S34, Calculate the grey relational coefficient: Relational coefficient ε ij The calculation formula is: In equation (3-9), ρ is the resolution coefficient, which is between 0 and 1. The resolution coefficient is introduced to reduce the influence of the extreme values ​​of the evaluation index on the calculation, thereby improving the discrimination of the correlation coefficient; the smaller the resolution coefficient, the greater the resolution. In general, the resolution coefficient is usually set to 0.5; S35, Calculate the grey relational degree: The correlation coefficient compares the degree of correlation between the sequences of each scheme and the optimal sequence on a certain indicator, but it cannot fully reflect the merits of the schemes. In order to grasp the overall degree of correlation between the sequences of schemes, it is necessary to weight and sum the correlation coefficients into a single value, namely the correlation degree p. Then, the correlation degree of the comparison sequence with respect to the optimal sequence is: In equation (3-10), w i The weight of the i-th indicator in the evaluation indicator system; S36, Ranking: Grey Relational Degree p j The larger the value, the closer the sequence of objects is to the optimal sequence, indicating that the j-th evaluated object is better. Therefore, based on the grey relational degree p... j The size will be used to sort the objects being evaluated.

7. The method for comprehensive evaluation of the competitiveness of charging stations according to claim 1, characterized in that, In step S4, the principle of TOPSIS analysis is to determine the standardized decision matrix by standardizing the evaluation index data and the weight of each index, and then find the optimal and worst solutions, i.e., the positive and negative ideal solutions, in the standardized decision matrix. Calculate the distance between each evaluation scheme and the positive and negative ideal solutions respectively, and finally use the relative closeness of each scheme to the optimal scheme to evaluate the quality of the scheme.

8. The method for comprehensive evaluation of the competitiveness of charging stations according to claim 7, characterized in that, The calculation steps for constructing the second comprehensive evaluation model of station competitiveness based on TOPSIS analysis include: S41, Establish the evaluation index matrix; S42, Standardization of evaluation indicators; S43, determine the index entropy weight; S44, Construct the decision matrix; S45, the positive and negative ideal solutions are determined by the optimal values ​​of each evaluation index in the decision matrix; S46, calculate the Euclidean distance between each evaluation object and the positive and negative ideal solutions; S47, calculate the relative proximity of each evaluation object to the optimal object.

9. The method for comprehensive evaluation of the competitiveness of charging stations according to claim 7, characterized in that, The specific calculation steps are as follows: S41, Establish the evaluation index matrix: First, assuming there are m evaluation objects and n evaluation indicators, establish the evaluation index matrix X, x ij This represents the value of the i-th indicator for the j-th evaluation object: S42, Standardization of evaluation indicators: Refer to formulas (2-2), (2-3), and (2-4) to standardize the evaluation indicator matrix. When the indicator value is as high as possible, it is considered a positive indicator, and its standardized formula is: When a smaller indicator value is better, then the indicator is considered an inverse indicator, and its standardized formula is: The closer the indicator value is to a certain value, the better the evaluation result; this value is denoted as... This indicator is a suitability indicator, and its standardized formula is: The normalized matrix R is obtained by processing matrix X according to the standardization formula: S43, Determine the entropy weight: Referring to the steps of the entropy weight method, calculate the entropy weight matrix W = (w1, w2, ..., w3) using formula (2-9). n ) T ,in S44, Constructing the decision matrix V: Constructing the decision matrix v based on evaluation indicators and entropy weights. ij =r ij ×w i : S45, determine the positive and negative ideal solutions Z based on the optimal values ​​of each evaluation index in the decision matrix. + and Z - : S46, Calculate the Euclidean distance between each evaluation object and the positive and negative ideal solutions. and S47, Calculate the relative proximity C between each evaluation object and the optimal object. j : In equation (4-11), C j The larger the value, the closer the j-th evaluation object is to the optimal object level.

Images